// Tile size: 3x4
	// Accumulators: 0-11
	// Col regs: ymm12-14
	// Row regs: ymm15

	vmovaps			ymm12,	[rax]
	vmovaps			ymm13,	[rax+32]
	vmovaps			ymm14,	[rax+64]

	vbroadcastss	ymm15,	dword ptr [rcx + 0]

	vfmadd231ps		ymm0,	ymm12, ymm15
	vfmadd231ps		ymm1,	ymm13, ymm15
	vfmadd231ps		ymm2,	ymm14, ymm15

	vbroadcastss	ymm15,	dword ptr [rcx + 4]

	vfmadd231ps		ymm3,	ymm12, ymm15
	vfmadd231ps		ymm4,	ymm13, ymm15
	vfmadd231ps		ymm5,	ymm14, ymm15

	vbroadcastss	ymm15,	dword ptr [rcx + 8]

	vfmadd231ps		ymm6,	ymm12, ymm15
	vfmadd231ps		ymm7,	ymm13, ymm15
	vfmadd231ps		ymm8,	ymm14, ymm15

	vbroadcastss	ymm15,	dword ptr [rcx + 12]

	vfmadd231ps		ymm9,	ymm12, ymm15
	vfmadd231ps		ymm10,	 ymm13, ymm15
	vfmadd231ps		ymm11,	 ymm14, ymm15